ANNUAL MOLT AND EGG PRODUCTION 83 



high, are rot classed as satisfactory breeders. It is desirable, therefore, to 

 find out whether the duration of annual molt is associated with second-year 

 Qgg productiouj. 



11. Duration of Molt versus Family Mortality. 



Records are available on 929 individuals showing an annual molt record. 

 Of this numfber 157 came from families with no mortality in the laying 

 houses. These 157 birds had a mean molt period of 82.25 days. Records on 

 772 birds from families with some mortality were tabulated against their 

 duration of molt and the following constants derived from a correlation 

 table: 



Number of birds _ 772 



Mean length of molt period — days 74.25 



Molt period standard deviation — days =!i30,.19 



Mean family mortality — per cent 25.35 



Mortality standard deviation- per cent ±14.50 



Coefficient of correlation -f .0828 ± .0241 



Correlation r a t i o . 1 898 



The mean length of molt period for the group of birds from families with 

 some mortality in the laying houses was 74.25 days as compared with a mean 

 mriiolt period of 82.25 days for the 157 individuals from families with no 

 mortality^ This suggests that the more vigorous birds did not molt quite as 

 rapidly as the less vigorous. 



Mean family mortality amounted to slightly over 25 per cent. The fact 

 should be noted in this connection that no culling is done after the pullets 

 are placed in the laying houses. Any decrease in nuniibers is, therefore, 

 brought about by the death of the birds. The range in mortality rate ex- 

 tended from 1 to 80 per cent. Evidently the flock was far from uniform in 

 vigor as measured by mortality rate. 



The coefficient of correlation appears to be insignificant. When the re- 

 gression of duration of molt period on mortality rate was tested by the X^ 

 method, the probability of linearity was .0915. The odds are, therefore, 

 greater than 10 to 1 against linearity, and the simple correlation coefficient 

 does not express the correlation desired. 



The correlation ratio of duration of molt to family mortality was .1898. 

 This differs fromi the coefficient by an amount 3.64 times as great as its 

 probable error. This constant correctly expresses the relationship desired 

 but it is of insignificant magnitude. It seems apparent that vigor and dur- 

 ation of molt were independent in the flock studied. 



12. Duration of Molt versus Second-Year Egg Production. 



A population of 541 birds furnished the records on molt duration and 

 second-year production. The second-year egg records began the day follow- 

 ing Uie close of the 365-day pullet year records and covered a period of 



